A train pantograph fault detection and alarm system
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-11
- Publication Date
- 2026-08-11
AI Technical Summary
[0003]然而传统的受电弓故障检测技术主要依赖于人工巡检和定期检测
[0049]本发明提供了一种列车受电弓故障检测与报警系统，系统包括：获取模块，用于获取目标列车的实时受电弓流媒体数据、列车实时运行数据、受电弓训练数据及相应时间的列车运行训练数据；训练模块，用于以所述受电弓训练数据及相应时间的列车运行训练数据，对初始建立的受电弓故障检测模型进行训练，得到目标受电弓故障检测模型；预测模块，用于将所述实时受电弓流媒体数据及所述列车实时运行数据输入所述目标受电弓故障检测模型，得到受电弓状态分类结果；报文生成模块，用于在所述受电弓状态分类结果为存在故障时，则基于相应的故障信息，生成故障检测结果报文；展示模块，用于展示所述故障检测结果报文中的故障区域、故障时间、故障类型、故障受电弓图像数据及其保存路径，并进行声光报警。从而司乘人员能够及时收到所展示的详细的故障信息，提升故障排查及处理效率，进而采取紧急措施，避免影响行车安全。
Smart Images
Figure CN122548359A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of rail transit fault prediction and health management technology, and in particular to a train pantograph fault detection and alarm system. Background Technology
[0002] As a crucial pillar of modern urban transportation, rail transit systems often experience equipment failures and operational interruptions due to their complexity and continuous high-load operation. To ensure the stability and safety of rail transit systems, fault prediction and health management systems have emerged and are widely used in actual operation. The pantograph, as a key component for obtaining power from trains, directly impacts train operation safety and efficiency due to its operational status. Therefore, real-time detection and alarm of pantograph faults are of paramount importance for preventing potential safety hazards and improving train operation safety.
[0003] However, traditional pantograph fault detection technologies mainly rely on manual inspections and periodic checks. This method not only consumes a lot of manpower and time, but also makes it difficult to achieve real-time and comprehensive monitoring of the pantograph's condition. More importantly, it cannot provide early warnings of potential fault risks, which could lead to serious safety accidents. Summary of the Invention
[0004] This disclosure provides a train pantograph fault detection and alarm system for real-time detection of train status and for real-time push of fault information when the train pantograph malfunctions.
[0005] In a first aspect, the present invention provides a train pantograph fault detection and alarm system, the system comprising:
[0006] The acquisition module is used to acquire real-time pantograph streaming media data, real-time train operation data, pantograph training data, and train operation training data for the corresponding time of the target train.
[0007] The training module is used to train the initially established pantograph fault detection model using the pantograph training data and the corresponding train operation training data to obtain the target pantograph fault detection model.
[0008] The prediction module is used to input the real-time pantograph streaming media data and the real-time train operation data into the target pantograph fault detection model to obtain the pantograph status classification result;
[0009] The message generation module is used to generate a fault detection result message based on the corresponding fault information when the pantograph status classification result indicates that a fault exists.
[0010] The display module is used to display the fault area, fault time, fault type, fault pantograph image data and its storage path in the fault detection result message, and to provide audible and visual alarms.
[0011] Optionally, the prediction module includes:
[0012] The parsing submodule is used to parse the real-time pantograph streaming media data to obtain pantograph image frame data;
[0013] The prediction submodule is used to process the pantograph image frame data using the target pantograph fault detection model and combine it with the real-time train operation data to obtain the pantograph status classification result.
[0014] Optionally, the pantograph training data includes: pantograph image frame training data and pantograph status labels; the training module includes:
[0015] The input submodule is used to input the pantograph image frame training data and the train operation training data into the pantograph fault detection model to obtain the corresponding pantograph state classification prediction results.
[0016] The training submodule is used to train the pantograph fault detection model based on the pantograph state classification prediction results and the pantograph state labels, so as to obtain the target pantograph fault detection model.
[0017] Optionally, the training submodule includes:
[0018] The training error determination unit is used to determine the training error based on the pantograph state classification prediction result and the pantograph state label;
[0019] An optimization unit is used to optimize the network parameters of the pantograph fault detection model based on the training error until the training termination condition is met, thereby obtaining the optimal network parameters.
[0020] The target model generation unit is used to generate the target pantograph fault detection model using the optimal network parameters.
[0021] Optionally, the display module includes:
[0022] The extraction submodule is used to extract and save the faulty pantograph image data corresponding to the frame number of the fault detection result message from the pantograph image frame data;
[0023] The annotation submodule is used to annotate the fault area in the fault pantograph image data based on the fault detection result message;
[0024] The display submodule is used to display the fault area, fault time, fault type, fault pantograph image data and its storage path in the fault detection result message, and to provide audible and visual alarms.
[0025] Secondly, the present invention provides a method for detecting and alarming pantograph faults in trains, the method comprising:
[0026] Acquire real-time pantograph streaming media data, real-time train operation data, pantograph training data, and train operation training data for the corresponding time period of the target train;
[0027] Using the pantograph training data and the corresponding train operation training data, the initially established pantograph fault detection model is trained to obtain the target pantograph fault detection model.
[0028] The real-time pantograph streaming media data and the real-time train operation data are input into the target pantograph fault detection model to obtain pantograph status classification results.
[0029] If the pantograph status classification result indicates a fault, a fault detection result message is generated based on the corresponding fault information.
[0030] The fault detection result message displays the fault area, fault time, fault type, fault pantograph image data and its storage path, and triggers an audible and visual alarm.
[0031] Optionally, the real-time pantograph streaming media data and the real-time train operation data are input into the target pantograph fault detection model to obtain pantograph status classification results, including:
[0032] The pantograph streaming media data is analyzed to obtain pantograph image frame data;
[0033] The pantograph image frame data is processed using the target pantograph fault detection model, and combined with the real-time train operation data to obtain the pantograph status classification result.
[0034] Optionally, the pantograph training data includes: pantograph image frame training data and pantograph status labels; using the pantograph training data and train operation training data at the corresponding time, the initially established pantograph fault detection model is trained to obtain the target pantograph fault detection model, including:
[0035] The pantograph image frame training data and the train operation training data are input into the pantograph fault detection model to obtain the corresponding pantograph status classification prediction results.
[0036] Based on the pantograph status classification prediction results and the pantograph status labels, the pantograph fault detection model is trained to obtain the target pantograph fault detection model.
[0037] Optionally, based on the pantograph state classification prediction results and the pantograph state labels, the pantograph fault detection model is trained to obtain the target pantograph fault detection model, including:
[0038] The training error is determined based on the pantograph state classification prediction results and the pantograph state labels;
[0039] Based on the training error, the network parameters of the pantograph fault detection model are optimized until the training termination condition is met, and the optimal network parameters are obtained.
[0040] The target pantograph fault detection model is generated using the optimal network parameters.
[0041] Optionally, the fault detection result message can be displayed, including the fault area, fault time, fault type, fault pantograph image data and its storage path, and an audible and visual alarm can be triggered, including:
[0042] Extract and save the faulty pantograph image data corresponding to the frame number of the fault detection result message from the pantograph image frame data;
[0043] In the image data of the faulty pantograph, the fault area is marked based on the fault detection result message;
[0044] The fault detection result message displays the fault area, fault time, fault type, fault pantograph image data and its storage path, and triggers an audible and visual alarm.
[0045] Thirdly, the present invention provides an electronic device including a processor and a memory, the memory storing computer-readable instructions that, when executed by the processor, perform the steps of the method provided in the second aspect above.
[0046] Fourthly, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the method provided in the second aspect above.
[0047] Fifthly, the present invention provides a computer program product comprising a computer program that, when executed by a processor, performs the steps of the method provided in the second aspect above.
[0048] As can be seen from the above technical solutions, the present invention has the following advantages:
[0049] This invention provides a train pantograph fault detection and alarm system. The system includes: an acquisition module for acquiring real-time pantograph streaming media data, real-time train operation data, pantograph training data, and corresponding time-based train operation training data of the target train; a training module for training an initially established pantograph fault detection model using the pantograph training data and the corresponding time-based train operation training data to obtain a target pantograph fault detection model; a prediction module for inputting the real-time pantograph streaming media data and the real-time train operation data into the target pantograph fault detection model to obtain a pantograph status classification result; a message generation module for generating a fault detection result message based on the corresponding fault information when the pantograph status classification result indicates a fault; and a display module for displaying the fault area, fault time, fault type, fault pantograph image data, and its storage path in the fault detection result message, and providing audible and visual alarms. This allows drivers and passengers to receive detailed fault information promptly, improving fault diagnosis and handling efficiency, and enabling them to take emergency measures to avoid affecting train operation safety. Attached Figure Description
[0050] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0051] Figure 1 This is a flowchart illustrating the steps of a train pantograph fault detection and alarm method according to an embodiment of the present invention.
[0052] Figure 2 This is a flowchart illustrating the steps of a second embodiment of the train pantograph fault detection and alarm method of the present invention.
[0053] Figure 3 This is a schematic diagram of the composition of a train pantograph fault detection and alarm device according to the present invention;
[0054] Figure 4 This is a structural block diagram of an embodiment of a train pantograph fault detection and alarm system according to the present invention. Detailed Implementation
[0055] This invention provides a train pantograph fault detection and alarm system, which is used to detect the train status in real time and push fault information in real time when the train pantograph fails.
[0056] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.
[0057] Example 1, please refer to Figure 1 , Figure 1 This is a flowchart illustrating the steps of a train pantograph fault detection and alarm method according to an embodiment of the present invention. The method includes:
[0058] Step S101: Obtain real-time pantograph streaming media data, real-time train operation data, pantograph training data, and train operation training data for the corresponding time period of the target train;
[0059] In this embodiment of the invention, high-definition cameras and sensors are installed in the pantograph area of the target train to collect images and operational status data of the pantograph.
[0060] In the specific implementation, in order to ensure the consistency of the timing of all collected data, a timestamp is added to each real-time pantograph streaming media data and real-time train operation data to facilitate accurate data matching in subsequent steps.
[0061] Step S102: Using the pantograph training data and the corresponding train operation training data, train the initially established pantograph fault detection model to obtain the target pantograph fault detection model.
[0062] In this invention, a pantograph fault detection model is first initialized. Then, the pantograph training data and train operation training data are preprocessed, including data cleaning, normalization, and denoising, to improve the model training effect. Next, useful features, such as pantograph image features, operating speed, and acceleration, are extracted from the pantograph and train operation training data to train the initially established pantograph fault detection model, adjusting model parameters and optimizing its fault detection capability. Finally, the trained pantograph fault detection model is evaluated to ensure it can accurately detect pantograph faults under various conditions.
[0063] Step S103: Input the real-time pantograph streaming media data and the real-time train operation data into the target pantograph fault detection model to obtain the pantograph status classification result;
[0064] In this embodiment of the invention, the input real-time pantograph streaming media data and real-time train operation data are first preprocessed, and then the preprocessed data is input into the target pantograph fault detection model to obtain pantograph status classification results (such as normal, fault, etc.).
[0065] Step S104: If the pantograph status classification result indicates a fault exists, a fault detection result message is generated based on the corresponding fault information.
[0066] In this embodiment of the invention, if the pantograph status classification result indicates a fault, detailed fault information, such as fault type, fault location, and fault severity, is extracted from the pantograph status classification result of the target pantograph fault detection model. Subsequently, the extracted fault information generates a fault detection result message, which includes the fault time, fault type, fault pantograph image data, and its storage path.
[0067] Step S105: Display the fault area, fault time, fault type, fault pantograph image data and its storage path in the fault detection result message, and issue an audible and visual alarm.
[0068] In this embodiment of the invention, the fault area, fault time, fault type, fault pantograph image data and its storage path in the fault detection result message are displayed on the interface of the central monitoring system for operators to view.
[0069] In practical applications, depending on the severity of the pantograph malfunction, a corresponding audible and visual alarm system can be triggered to remind operators to handle the malfunction in a timely manner.
[0070] This invention provides a method for detecting and alarming pantograph faults in trains, comprising: acquiring real-time pantograph streaming media data, real-time train operation data, pantograph training data, and train operation training data at corresponding times for a target train; training an initially established pantograph fault detection model using the pantograph training data and the train operation training data at corresponding times to obtain a target pantograph fault detection model; inputting the real-time pantograph streaming media data and the real-time train operation data into the target pantograph fault detection model to obtain a pantograph status classification result; if the pantograph status classification result indicates a fault, generating a fault detection result message based on the corresponding fault information; displaying the fault area, fault time, fault type, fault pantograph image data, and its storage path in the fault detection result message, and triggering an audible and visual alarm. This allows drivers and passengers to receive detailed fault information promptly, improving fault diagnosis and handling efficiency, and enabling them to take emergency measures to avoid affecting train operation safety.
[0071] Example 2, please refer to Figure 2 , Figure 2This is a flowchart illustrating a second embodiment of the train pantograph fault detection and alarm method of the present invention. The steps include:
[0072] Step S201: Obtain real-time pantograph streaming media data, real-time train operation data, pantograph training data, and train operation training data for the corresponding time of the target train; the pantograph training data includes: pantograph image frame training data and pantograph status labels;
[0073] In this embodiment of the application, the train pantograph fault detection method is applied to, for example... Figure 3 The train pantograph fault detection device shown includes a server 1, a storage board 2, an Ethernet board 3, and a power supply board 4. The server 1 is connected to the pantograph camera 5 and the Ethernet board 3 via Ethernet, and to the storage board 2 via a PCIe interface. The Ethernet board 3 communicates with the on-board monitoring screen 6 via Ethernet. The power supply board 4 is externally connected to a 110V power supply 7.
[0074] Server 1 is used to perform fault detection and alarm for the train pantograph, including acquiring real-time pantograph streaming media data, parsing real-time train operation data, detecting pantograph faults, and generating fault detection result messages and issuing audible and visual alarms when faults exist. Server 1 includes a target pantograph fault detection model, service algorithms, and a streaming media server.
[0075] Storage board 2 is a high-capacity storage board, connected to a hard drive for long-term storage of faulty pantograph image data and fault videos.
[0076] Ethernet board 3 is used to enable Ethernet communication between server 1 and vehicle monitoring screen 6, as well as between various boards in the system.
[0077] Power board 4 is used to provide DC 12V or DC 5V power to the various components of the system.
[0078] The onboard monitoring screen 6 is equipped with pantograph monitoring software to monitor the pantograph and analyze the operating status of the train in real time. It supports functions such as fault pop-up alarm, fault file display, and fault parameter setting.
[0079] In the specific implementation, after the train is powered on, server 1 sends a request to the video server of pantograph camera 6 via the RTSP protocol to obtain real-time pantograph streaming media data and create a video buffer queue. The queue length is the buffer duration multiplied by the frame rate, and then it begins to receive real-time pantograph streaming media data.
[0080] Step S202: Input the pantograph image frame training data and the train operation training data into the pantograph fault detection model to obtain the corresponding pantograph state classification prediction results;
[0081] Step S203: Based on the pantograph status classification prediction results and the pantograph status labels, train the pantograph fault detection model to obtain the target pantograph fault detection model;
[0082] In this embodiment, the training error is first determined based on the pantograph state classification prediction results and pantograph state labels. Then, based on the training error, the network parameters of the pantograph fault detection model are optimized until the training termination condition is met, and the optimal network parameters are obtained. Finally, the optimal network parameters are used to generate the target pantograph fault detection model.
[0083] Step S204: Analyze the real-time pantograph streaming media data to obtain pantograph image frame data;
[0084] In this embodiment, server 1 parses the real-time pantograph streaming media data to obtain pantograph image frame data, and stores the color information of each pixel in BGR format. Server 1 then inputs the pantograph image frame data into the pantograph fault detection model, and simultaneously writes the pantograph image frame data, encoded in JPEG format and along with the corresponding frame number, into a cache queue using a FIFO elimination algorithm. Each time pantograph image frame data is successfully acquired, the system time is recorded. If the difference between the current system time and the last successful acquisition time is greater than 5 seconds, a reconnection operation is performed to the video server of the pantograph camera 5.
[0085] Step S205: The pantograph image frame data is processed using the target pantograph fault detection model, and combined with the real-time train operation data to obtain the pantograph status classification result;
[0086] In this embodiment, by collecting a large amount of pantograph training data and corresponding train operation training data, computer vision is used for image enhancement and feature extraction. A deep learning network model is used to train the extracted features over a long period of time to obtain a target pantograph fault detection model, which can classify the input pantograph image frame data and determine whether there is a fault.
[0087] In the specific implementation, server 1 uses the built-in data preprocessing program to parse the real-time pantograph streaming media data. Then, it processes the pantograph image frame data obtained from the preprocessing in real time through the target pantograph fault detection model. Combining common parameters such as pantograph lifting status and vehicle speed, it automatically completes the detection of more than ten kinds of faults, such as pantograph arcing, foreign objects hanging on the pantograph, dirty base image, abnormal pantograph structure, and abnormal surge arrester.
[0088] Step S206: If the pantograph status classification result indicates a fault, then a fault detection result message is generated based on the corresponding fault information.
[0089] In this embodiment of the application, if the classification result indicates that a fault exists, the target pantograph fault detection model writes parameters such as spark size, pantograph deflection angle, and fault type into the fault detection result message.
[0090] In the specific implementation, after receiving the fault detection result message, server 1 fills the fault result message with the fault type, alarm level, the fault pantograph image data and its storage path, and sends it to the vehicle monitoring screen 6 via TCP.
[0091] It should be noted that the detection algorithm and detection type are not limited to those described above. The fault detection algorithm can be continuously optimized by updating the software via Ethernet, depending on the needs of upgrading the target pantograph fault detection model or changes in the application environment.
[0092] Step S207: Extract and save the faulty pantograph image data corresponding to the frame number of the fault detection result message from the pantograph image frame data;
[0093] In this embodiment of the invention, after the server 1 parses the detection result message, it retrieves the corresponding frame data from the video cache queue according to the fault image frame number and fault parameters, marks the fault area in the image, and generates an image that is temporarily stored in the storage board 2.
[0094] In the specific implementation, server 1 retrieves one minute of pantograph image data from the cache queue and saves it as the video one minute before the fault. After waiting for one minute, it retrieves two minutes of pantograph image data from the video cache queue and saves it as the video one minute before and after the fault. For each fault file generated, server 1 renames the file according to the format "date_carriage number_position number_time_train number.file type", and then transfers the file to storage board 2 for long-term storage via anonymous FTP. The storage path format is "date / file type / carriage number + position number / alarm type / fault type". Maintenance personnel can download fault files of a specified time and type via anonymous FTP, facilitating fault diagnosis and location.
[0095] Storage board 2 features a hot-swappable hard drive design, allowing for configuration of hard drives with varying storage capacities to meet project requirements. Maintenance personnel can remove the hard drive containing faulty files and take it to the ground for data retrieval and fault analysis.
[0096] Step S208: Mark the fault area in the fault pantograph image data based on the fault detection result message;
[0097] Step S209: Display the fault area, fault time, fault type, fault pantograph image data and its storage path in the fault detection result message, and issue an audible and visual alarm.
[0098] In this embodiment of the invention, the on-board monitoring screen 6 displays information such as the fault area, fault time, fault type, fault pantograph image data, and its storage path on the pantograph fault monitoring interface. Maintenance personnel can manually click on images to transmit the fault pantograph image data locally for display; they can also manually click on videos to send a fault video stream request to server 1, subsequently starting playback of the fault video, supporting functions such as dragging the fault video progress bar and modifying the playback speed.
[0099] In addition, maintenance personnel can use the historical fault query function to set parameters such as fault time range, fault level, and fault type. During operation, a remote search message is sent to server 1. Server 1 retrieves the fault file path that meets the conditions from storage board 2, writes the fault file path into the remote search response message, and sends it to the vehicle monitoring screen 6 via TCP for fault display.
[0100] This invention discloses a method for detecting and alarming pantograph faults on trains, comprising: acquiring real-time pantograph streaming media data, real-time train operation data, pantograph training data, and train operation training data at corresponding times for a target train; training an initially established pantograph fault detection model using the pantograph training data and the train operation training data at corresponding times to obtain a target pantograph fault detection model; inputting the real-time pantograph streaming media data and the real-time train operation data into the target pantograph fault detection model to obtain a pantograph status classification result; if the pantograph status classification result indicates a fault, generating a fault detection result message based on the corresponding fault information; displaying the fault area, fault time, fault type, fault pantograph image data and its storage path in the fault detection result message, and triggering an audible and visual alarm. This method not only allows drivers and passengers to receive detailed fault information promptly, improving fault diagnosis and handling efficiency and enabling emergency measures to be taken to avoid affecting train operation safety, but also adds functions such as fault image storage and annotation, and remote fault query, enhancing post-fault analysis capabilities.
[0101] Example 3, please refer to Figure 4 , Figure 4 This is a structural block diagram of an embodiment of a train pantograph fault detection and alarm system according to the present invention. The device includes:
[0102] The acquisition module 301 is used to acquire real-time pantograph streaming media data, real-time train operation data, pantograph training data and train operation training data for the corresponding time of the target train.
[0103] Training module 302 is used to train the initially established pantograph fault detection model using the pantograph training data and the train operation training data of the corresponding time, so as to obtain the target pantograph fault detection model.
[0104] Prediction module 303 is used to input the real-time pantograph streaming media data and the real-time train operation data into the target pantograph fault detection model to obtain pantograph status classification results;
[0105] The message generation module 304 is used to generate a fault detection result message based on the corresponding fault information when the pantograph status classification result indicates that a fault exists.
[0106] The display module 305 is used to display the fault area, fault time, fault type, fault pantograph image data and its storage path in the fault detection result message, and to provide audible and visual alarms.
[0107] In an optional embodiment, the prediction module 303 includes:
[0108] The parsing submodule is used to parse the real-time pantograph streaming media data to obtain pantograph image frame data;
[0109] The prediction submodule is used to process the pantograph image frame data using the target pantograph fault detection model and combine it with the real-time train operation data to obtain the pantograph status classification result.
[0110] In an optional embodiment, the pantograph training data includes: pantograph image frame training data and pantograph status labels; the training module 302 includes:
[0111] The input submodule is used to input the pantograph image frame training data and the train operation training data into the pantograph fault detection model to obtain the corresponding pantograph state classification prediction results.
[0112] The training submodule is used to train the pantograph fault detection model based on the pantograph state classification prediction results and the pantograph state labels, so as to obtain the target pantograph fault detection model.
[0113] In an optional embodiment, the training submodule includes:
[0114] The training error determination unit is used to determine the training error based on the pantograph state classification prediction result and the pantograph state label;
[0115] An optimization unit is used to optimize the network parameters of the pantograph fault detection model based on the training error until the training termination condition is met, thereby obtaining the optimal network parameters.
[0116] The target model generation unit is used to generate the target pantograph fault detection model using the optimal network parameters.
[0117] In an optional embodiment, the display module 305 includes:
[0118] The extraction submodule is used to extract and save the faulty pantograph image data corresponding to the frame number of the fault detection result message from the pantograph image frame data;
[0119] The annotation submodule is used to annotate the fault area in the fault pantograph image data based on the fault detection result message;
[0120] The display submodule is used to display the fault area, fault time, fault type, fault pantograph image data and its storage path in the fault detection result message, and to provide audible and visual alarms.
[0121] Example 4: This embodiment of the invention also provides an electronic device, including a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor performs the steps of a train pantograph fault detection and alarm method according to any embodiment.
[0122] Example 5: This embodiment of the invention also provides a computer storage medium storing a computer program thereon. When the computer program is executed by the processor, it implements the steps of a train pantograph fault detection and alarm method according to any embodiment.
[0123] Example 6: This embodiment of the invention also provides a computer program product, on which a computer program is stored. When the computer program is executed by the processor, it implements the steps of a train pantograph fault detection and alarm method according to any embodiment.
[0124] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.
[0125] In the several embodiments provided in this application, it should be understood that the methods, apparatuses, electronic devices, and storage media disclosed in this invention can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.
[0126] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0127] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0128] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned readable storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0129] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A pantograph fault detection and alarm system for a train, characterized in that, include: The acquisition module is used to acquire real-time pantograph streaming media data, real-time train operation data, pantograph training data, and train operation training data for the corresponding time of the target train. The training module is used to train the initially established pantograph fault detection model using the pantograph training data and the train operation training data of the corresponding time, so as to obtain the target pantograph fault detection model. The prediction module is used to input the real-time pantograph streaming media data and the real-time train operation data into the target pantograph fault detection model to obtain the pantograph status classification result; The message generation module is used to generate a fault detection result message based on the corresponding fault information when the pantograph status classification result indicates that a fault exists. The display module is used to display the fault area, fault time, fault type, fault pantograph image data and its storage path in the fault detection result message, and to provide audible and visual alarms.
2. The pantograph fault detection and alarm system of claim 1, wherein The prediction module includes: The parsing submodule is used to parse the real-time pantograph streaming media data to obtain pantograph image frame data; The prediction submodule is used to process the pantograph image frame data using the target pantograph fault detection model and combine it with the real-time train operation data to obtain the pantograph status classification result.
3. The pantograph fault detection and alarm system of claim 2, wherein The pantograph training data includes: pantograph image frame training data and pantograph status labels; the training module includes: The input submodule is used to input the pantograph image frame training data and the train operation training data into the pantograph fault detection model to obtain the corresponding pantograph state classification prediction results. The training submodule is used to train the pantograph fault detection model based on the pantograph state classification prediction results and the pantograph state labels, so as to obtain the target pantograph fault detection model.
4. The pantograph fault detection and alarm system of claim 3, wherein The training submodule includes: The training error determination unit is used to determine the training error based on the pantograph state classification prediction result and the pantograph state label; An optimization unit is used to optimize the network parameters of the pantograph fault detection model based on the training error until the training termination condition is met, thereby obtaining the optimal network parameters. The target model generation unit is used to generate the target pantograph fault detection model using the optimal network parameters.
5. The pantograph fault detection and alarm system of claim 2, wherein The display module includes: The extraction submodule is used to extract and save the faulty pantograph image data corresponding to the frame number of the fault detection result message from the pantograph image frame data; The annotation submodule is used to annotate the fault area in the fault pantograph image data based on the fault detection result message; The display submodule is used to display the fault area, fault time, fault type, fault pantograph image data and its storage path in the fault detection result message, and to provide audible and visual alarms.
6. A method for detecting and alarming the failure of a train pantograph, characterized in that, The methods include: Acquire real-time pantograph streaming media data, real-time train operation data, pantograph training data, and train operation training data for the corresponding time period of the target train; Using the pantograph training data and the corresponding train operation training data, the initially established pantograph fault detection model is trained to obtain the target pantograph fault detection model. The real-time pantograph streaming media data and the real-time train operation data are input into the target pantograph fault detection model to obtain pantograph status classification results. If the pantograph status classification result indicates a fault, a fault detection result message is generated based on the corresponding fault information. The fault detection result message displays the fault area, fault time, fault type, fault pantograph image data and its storage path, and triggers an audible and visual alarm.
7. The method of claim 6, wherein the method further comprises: The real-time pantograph streaming media data and the real-time train operation data are input into the target pantograph fault detection model to obtain pantograph status classification results, including: The pantograph streaming media data is analyzed to obtain pantograph image frame data; The pantograph image frame data is processed using the target pantograph fault detection model, and combined with the real-time train operation data to obtain the pantograph status classification result.
8. An electronic device, comprising: It includes a processor and a memory, the memory storing computer-readable instructions that, when executed by the processor, perform the method as described in any one of claims 6-7.
9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it performs the method as described in any one of claims 6-7.
10. A computer program product comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 6-7.